282 EREMOPHYTES sect, xi 



close carpet. Both these features serve to distinguish grass-steppe 

 from the alhed formation, meadow, whose close vegetation consists of 

 fresh-green, soft-leaved, and broad-leaved grasses and other perennial 

 herbs. On the other hand, vegetation clothing grass-steppe is closer and 

 taller than that of desert. 



There are usually two periods of rest, one caused by drought in 

 summer, the other by cold in winter. 



The soil varies widely in nature. But characteristic of steppe is the 

 exceedingly common occurrence of loess. In places where grass-steppe 

 bounds more open steppe, or even desert, the drifting dust is detained 

 by the denser vegetation of grass-steppe, where it consequently comes 

 to rest. It is in this manner, according to Richthofen, that the thick 

 strata of loess have been deposited in northern China. But grass-steppe 

 can itself give rise to loess, as in summer there are, between the grasses, 

 many bare patches, from which the wind can raise the dust and deposit 

 it elsewhere. A particular type of loess is the ' black earth ', which occurs 

 in Russian and Turanian steppes, in the American prairies, in Morocco, 

 and elsewhere. It abounds in humus.^ Loess is very rich in nutritive 

 salts, and 'black earth' is specially so. 



Grass-steppe in the Old World. 

 South-eastern Russia. 



The steppes of southern Russia and the pusztas of Hungary have 

 vegetation that is identical in oecology and to some extent in flora. 



The problem of the origin of steppes has given rise to a considerable 

 literature ; von Baer, Ruprecht, Dokuchayev, and Tanfiljew are of opinion 

 that they were always steppes, while Pallas and Pahmpsestow beheve that 

 they developed from ruined forests. 



There is also a discussion as to whether the distribution of forest 

 and steppe is due to climate or to soil. Von Baer suggests that prolonged 

 dryness is responsible for the lack of trees in steppe ; Middendorff's 

 opinion was that hot, dry winds were responsible for this ; Beketow, 

 Dokuchayev, the geologist, and Tanfiljew, the phyto-geographer, regard 

 salinity of soil as the reason why forests have not advanced on steppe-soil. 

 And Tanfiljew points out that forest actually does advance when the 

 soil is gradually elutriated. On steppe-regions forest is met with especially 

 on the ridges of watersheds, partly because these ridges are more elutriated 

 than other spots. 



But climatologists ^ all agree that the climate of steppe is in reahty 

 drier than that of forest. In particular, the frequency of rain is small. 

 Rain falls in few but heavy downpours, so that the main mass of water 

 runs away superficially without penetrating the soil. In regions where 

 grass-steppe prevails tree-growth appears by no means to be absolutely 

 excluded, but only rendered difficult by climatic causes. In Caucasia, 

 Radde^ noted that on the banks of rivers traversing the steppes trees 

 in the riparian forests are stunted, and that many of their leaves dry up 

 in summer. In these regions forest is confined to edaphically favourable 

 localities. Here it occurs in river valleys, but likewise on the less sahne 



* Cp. Kostytscheff, 1890; Albert, 1907. 



" Woikof. Hann, and Koppen. See Rikli, 1907 i, p. 107. ^ Radde, 1899. 



